Academic Intelligence · Curated Daily

Explore the Frontier of Global Academia

AcademicHub aggregates real-time literature from top journals and preprint platforms. Build your personal research radar and let large language models compile cross-disciplinary analysis briefings automatically.

01.
PLOS Computational Biology 2026-06-01

BeetleAtlas 2: An enhanced <i>Tribolium castaneum</i> web resource for tissue and developmental transcriptomics allowing refinement of gene predictions

by David P. Leader, Muhammad T. Naseem, Janina L. Rinke, Kenneth Veland Halberg BeetleAtlas is an online resource for tissue- and stage-specific transcriptomics in the red flour beetle, Tribolium castaneum. On updating from the original Tcas5.2 genome assembly to the more recent improved icTriCast1.1 genome assembly it became evident that there were major discrepancies between the gene models of the two genome annotations in use: the OGS3 and the NCBI gene sets. As neither was clearly superior we implemented a new design in BeetleAtlas 2 (beetleatlas.org) comprising two parallel ‘modes’ — one incorporating results using the NCBI gene models and a second incorporating those using the OGS3 gene models. This allows direct comparison where equivalent gene models exist: 50–57% of cases. To aid resolution of discrepancies between the two gene model sets and verification of results, gene models are linked to a custom visualization of RNA-seq read coverage of the genome in the UCSC Genome Browser. This displays reads from 22 tissues and life stages superimposed on the icTriCast1.1 genome assembly. Reference tracks show the NCBI gene models, the OGS3 gene models after translation of their coordinates from the Tcas5.2 assembly, and 1050 discontinued NCBI gene models from the previous assembly after a similar transfer of coordinates. We document various situations in which distinct patterns of expression of the tissues can be used to confirm and extend correlations between the two gene sets, resolve discrepancies between them, make corrections and identify putative genes or exons absent from the current gene sets. BeetleAtlas 2 allows those involved in Tribolium research to avoid the pitfalls inherent in incorrect gene models when planning experiments on specific genes and interpreting the results. It also demonstrates how BeetleAtlas 2 might play an important role in establishing a revised gene set for Tribolium castaneum in the future.

02.
arXiv (math.PR) 2026-06-18

Very large cliques in a scale-free random graph

arXiv:2606.18722v1 Announce Type: new Abstract: In this short article we consider a preferential attachment random graph model with edge steps, studied by Alves, Ribeiro and Sanchis. Starting with an initial graph $\mathbb{G}_1$ formed by a vertex with a self-loop attached to it, the model evolves as follows. At every subsequent (discrete) time step, either with probability $p$ we add a vertex to the graph and connect it to exactly one of the older vertices selected with probability proportional to its degree, or with probability $1-p$ we add one edge between two existing vertices, both selected (independently) with probability proportional to their degrees. Let $\omega(\mathbb{G})$ be the clique number of a graph $\mathbb{G}$, i.e.\ the number of vertices in a largest complete subgraph of $\mathbb{G}_{}$. Alves, Ribeiro and Sanchis showed that, for any given $\varepsilon>0$, we have $\omega(\mathbb{G}_{2t})\geq t^{\frac{1-p}{2-p}(1-\varepsilon)}$ with high probability (i.e.\ with probability tending to $1$ as $t\rightarrow \infty$). Here we strengthen this bound by showing that, for any function $f:\mathbb{N}\mapsto \mathbb{N}$ that satisfies $f(t)\rightarrow \infty$ as $t\rightarrow \infty$, with high probability \[\omega(\mathbb{G}_{2t}) = \Omega\left(t^{\frac{1-p}{2-p}}\Big(\log^{\frac{1}{2-p}}(t)f(t)\Big)^{-1}\right).\]

03.
arXiv (CS.AI) 2026-06-11

ATLAS: Active Theory Learning for Automated Science

arXiv:2606.12386v1 Announce Type: cross Abstract: Advancing scientific understanding through mechanistic modeling requires posing the right experimental questions to yield maximally informative data. To automate this pursuit within cognitive science, we introduce ATLAS (Active Theory Learning for Automated Science), an active learning framework for the data-driven discovery of interpretable behavioral models. ATLAS iterates between generating mechanistic hypotheses–instantiated as a diverse ensemble of sparse neural networks (Disentangled RNNs)–and designing experiments that optimally distinguish between them. We test this approach on the problem of recovering reinforcement learning agents from their behavior in bandit tasks. ATLAS designs varied sequences of qualitatively novel experiments with temporal structure tailored to underlying agent characteristics. The models trained on these experiments are evaluated against a comprehensive set of metrics for mechanistic modeling that capture behavioral, structural, and computational similarity. ATLAS achieves a 5-10x improvement in sample efficiency across all metrics compared to random experimentation, and its performance is further validated against expert-designed experiments derived from literature. These in silico results showcase ATLAS's potential to accelerate human-interpretable insights in cognitive science and other domains where scientific inquiry relies on discovering mechanistic models.

04.
arXiv (CS.CL) 2026-06-15

Creative Integration: A Decidable Criterion of Creativity

"Integrative" solutions are widely praised but rarely defined: we lack an operational way to tell a genuine integration – one that makes the world cheaper to describe – from a tidy re-description. Building on the lineage that treats creativity and intelligence as compression, we give such a criterion for creative integration (CI): the resolution of a real conflict between A and B is CI if and only if, under a fixed description language, the description length strictly shrinks (C = L_pre/L_post > 1), with the reduction located in the conflict itself. We make the judgment decidable through four binary, conjunctive gates, and we fix its extension through a taxonomy of pseudo-integration that names and rejects the look-alikes. We back the criterion with a curated, multi-domain corpus and – crucially – validate it not by human inter-rater agreement but by four falsifiable tests it could fail: an independent computational check, discrimination against hard negatives, out-of-sample prediction, and description-language robustness; all pass with margin. The contribution is not "creativity is compression" but its decidability, discrimination, and corpus: on this account, what makes a move genuinely creative – rather than merely novel – is that it compresses a conflict, with novelty and value as downstream symptoms; whether all creativity is so constituted we state as an explicit conjecture. We claim only the sign of C-1; we judge, not generate. The result is a citable primitive for a broader program.

05.
arXiv (CS.AI) 2026-06-19

Hierarchical Control in Multi-Agent Games: LLM-based Planning and RL Execution

arXiv:2606.20014v1 Announce Type: cross Abstract: Reinforcement learning (RL) has achieved strong performance in sequential decision-making, yet scaling to complex multi-agent environments remains challenging due to sparse rewards, large state-action spaces, and the difficulty of learning coordinated strategies. We propose a hierarchical architecture where a pretrained large language model (LLM) acts as a centralized strategic controller that selects among specialized RL skill policies for a team of agents, while RL policies handle reactive low-level execution. We evaluate this hybrid system in a competitive 2v2 King of the Hill environment against behavior tree (BT) and ``Flat'' RL (end-to-end training without skill decomposition) baselines. The LLM+RL system achieves task performance statistically equivalent to hand-crafted BT (46.4\% vs 51.5\% win rate, $p=0.103$) while both significantly outperform Flat RL trained without skill decomposition. A user study ($n=15$) reveals that 60\% of participants perceive LLM+RL agents as the most human-like ($p=0.027$), citing behavioral adaptability and tactical variability. These results demonstrate that pretrained LLM reasoning can effectively orchestrate pretrained RL skills, achieving competitive multi-agent coordination and superior perceived believability without manual rule engineering.

06.
bioRxiv (Bioinfo) 2026-06-11

A Deep Hypergraph Learning Model for Predicting Antimicrobial Combination Effects Across Bacterial Targets

Antimicrobial resistance (AMR) creates an urgent need for efficient strategies to identify effective antibacterial combinations. Combination therapy, including antimicrobial peptides (AMPs) paired with conventional antibiotics, is a promising approach, but exhaustive experimental screening across drug pairs and bacterial targets is impractical. This study introduces a hybrid GCN-based hypergraph neural network (HGNN) for predicting antimicrobial-agent combination outcomes against bacterial targets. Each antimicrobial-agent-antimicrobial-agent-bacterium triplet is represented as a ternary hyperedge, enabling the model to learn context-dependent interaction patterns. The framework integrates SMILES-derived molecular graph embeddings for antimicrobial agents, including conventional antibiotics and AMPs, with taxonomy-derived bacterial representations. The prediction task was formulated as a three-class classification problem: synergy, antagonism, and non-interaction. The non-interaction class included experimentally verified indifferent records and synthetic presumed non-interaction triplets generated by negative sampling. Model development used drug-pair-grouped splitting, five-fold grouped cross-validation within the training/validation partition, and final evaluation on a held-out test set. On the held-out three-class test set, the selected GCN-based HGNN achieved an accuracy of 0.83, weighted F1-score of 0.84, macro F1-score of 0.80, and ROC-AUC of 0.95. Per-class evaluation showed accuracies of 0.80 for synergy, 0.92 for antagonism, and 0.85 for non-interaction. Pair-type analysis showed strong performance across AMP-AMP, AMP-conventional antibiotic, and conventional antibiotic-conventional antibiotic combinations. These findings suggest that hypergraph-based representation learning can support computational prioritization of antimicrobial combinations for experimental follow-up. Further studies will be needed to improve model interpretability and to perform prospective validation of predicted synergistic combinations.

07.
arXiv (CS.AI) 2026-06-11

Feature-Aligned Speech Watermarking for Robustness to Reconstruction Distortions

arXiv:2606.11828v1 Announce Type: cross Abstract: Audio watermarking aims to embed identifiable information into audio while remaining imperceptible. Existing methods adopt high-fidelity, low-energy designs to preserve perceptual quality, but the resulting watermarks lack robustness under suppression by speech reconstruction models. Improving robustness is challenging due to the inherent robustness-fidelity trade-off in existing designs, where increasing watermark energy improves robustness but reduces fidelity. To address this problem, we propose a feature-aligned watermarking method that aligns the watermark with the original speech feature distribution, allowing higher watermark energy to improve robustness while preserving imperceptibility. We use a pretrained speech codec to generate a pseudo-speech watermark and fuse it into the spectrogram of the input audio, with VAD loss and perceptual losses guiding embedding within voiced regions. Experiments show that our method maintains imperceptibility comparable to existing approaches while substantially improving robustness under both seen and unseen speech reconstruction models.

08.
arXiv (CS.LG) 2026-06-16

Finite-Width Neural Tangent Kernels from Feynman Diagrams

arXiv:2508.11522v4 Announce Type: replace Abstract: Neural tangent kernels (NTKs) are a powerful tool for analyzing deep, non-linear neural networks. In the infinite-width limit, NTKs can easily be computed for most common architectures, yielding full analytic control over the training dynamics. However, at infinite width, important properties of training such as NTK evolution or feature learning are absent. Nevertheless, finite width effects can be included by computing corrections to the Gaussian statistics at infinite width. We introduce Feynman diagrams for computing finite-width corrections to NTK statistics. These dramatically simplify the necessary algebraic manipulations and enable the computation of layer-wise recursion relations for arbitrary statistics involving preactivations, NTKs and certain higher-derivative tensors (dNTK and ddNTK) required to predict the training dynamics at leading order. We demonstrate the feasibility of our framework by extending stability results for deep networks from preactivations to NTKs and proving the absence of finite-width corrections for scale-invariant nonlinearities such as ReLU on the diagonal of the Gram matrix of the NTK. We numerically implement the complete set of equations necessary to compute the first-order corrections for arbitrary inputs and demonstrate that the results follow the statistics of sampled neural networks for widths $n\gtrsim 20$.

09.
arXiv (CS.AI) 2026-06-19

What Do Safety-Aligned LLMs Learn From Mixed Compliance Demonstrations?

arXiv:2606.20508v1 Announce Type: new Abstract: Prior work has shown that in-context demonstrations can jailbreak language models, but it remains unclear how models interpret different types of compliance demonstrations. We study this by mixing benign compliance demonstrations (non-harmful request, helpful response) with harmful compliance demonstrations (harmful request, helpful response) and testing three hypotheses about how demonstration composition drives harmful compliance. Across four models, we find that benign and harmful demonstrations are not interchangeable: benign demonstrations can either reduce or increase harmful compliance depending on the model. We further show that preference optimization is the critical training stage that prevents benign demonstrations from increasing harmful compliance, that demonstration ordering exhibits strong recency bias, and that models differ in how refusal interacts with in-context learning: some adopt demonstrated formatting even when refusing, while others override all in-context signals upon refusal. Taken together, this work moves beyond showing that demonstration-based jailbreaking works to characterizing how it works: what models extract from compliance demonstrations depends on demonstration content, ordering, and training methodology.

10.
arXiv (CS.LG) 2026-06-16

Causal-Privacy Audit Workflow for Synthetic and Distilled Data in Dropout Support

arXiv:2606.15940v1 Announce Type: new Abstract: Synthetic and distilled student data are increasingly used to enable privacy-conscious learning analytics, yet their suitability for decision-facing institutional support remains uncertain. In dropout support, generated data must preserve not only predictive utility or distributional resemblance, but also the financial-status evidence used to guide advising, payment-plan assistance, and scholarship-related decisions. Method: This study introduces CaP-Eval, a decision-facing causal-privacy audit workflow for evaluating generated student data under a fixed estimand, timing-aware adjustment design, estimator set, and empirical privacy-governance screen. The workflow compares original, distilled, adversarial synthetic, statistical synthetic, and DPGNet privacy-oriented generated data on predictive utility, treatment-effect fidelity, robustness to alternative estimators, and local training-record proximity. Results: DPGNet and distilled data preserved the original financial-status treatment-effect structure more reliably than the adversarial and Gaussian Copula baselines. DPGNet preserved full direction and rank agreement across epsilon levels; epsilon = 10 produced the smallest non-original IPW and DML deviations, while epsilon = 1 and epsilon = 5 amplified several financial-status contrasts. Distilled data remained highly faithful but retained the strongest local training-record proximity signal. TabularGNet preserved qualitative directions with moderate attenuation, and Gaussian Copula compressed effect magnitudes. Conclusions: Predictive utility, privacy orientation, empirical disclosure signals, and causal fidelity diverged; generated student data require joint audits of direction, magnitude, overlap, and release-governance risk before decision use.

11.
arXiv (CS.LG) 2026-06-15

Riemannian Metric Matching for Scalable Geometric Modeling of Distributions

arXiv:2606.14334v1 Announce Type: new Abstract: High-dimensional datasets often concentrate near low-dimensional structures, but estimating their geometry from samples typically relies on graphs and kernels that scale poorly with dataset size and dimension. We propose Riemannian metric matching: a denoising probabilistic framework for learning the Riemannian geometry of data using neural networks. Specifically, we learn the carré du champ operator, which, using diffusion geometry, gives us access to the Riemannian geometry toolkit for downstream machine learning and statistical tasks. Our key observation is that the carré du champ operator can be formulated as a conditional expectation over random perturbations of the data, which can be exploited for sample-wise training and constant cost, amortized inference without explicit kernel construction. Empirically, metric matching rivals or improves the accuracy of $k$-NN-based diffusion geometry estimators, while enabling amortized inference that is up to $400\times$ faster, and supports graph-free geometric analysis on high-dimensional images where nearest neighbors break down.

12.
arXiv (CS.AI) 2026-06-16

LaWAM: Latent World Action Models for Efficient Dynamics-Aware Robot Policies

arXiv:2606.15768v1 Announce Type: cross Abstract: Vision-Language-Action models (VLAs) leverage large-scale vision-language pretraining for semantic robot control, but often lack explicit foresight into how robot actions change the scene. World-Action Models (WAMs) address this limitation by conditioning policies on predicted futures, yet existing approaches typically rely on computationally expensive video generation with substantial pixel-level redundancy. We present LaWAM, a Latent World Action Model that exposes predictive dynamics to robot policies through compact latent visual subgoals instead of reconstructed future video. At the core of LaWAM is a latent-action-conditioned Latent World Model (LaWM). We obtain LaWM by training a latent action model in the latent space of a pretrained vision foundation model and repurposing its forward decoder to predict future observation features for scene evolution. LaWAM then conditions action generation on these predicted latent visual subgoals to enable dynamics-aware robot control. LaWAM achieves state-of-the-art or competitive success rates (SRs) across LIBERO (98.6% SR), RoboTwin (91.22% SR), and real-world manipulation tasks while retaining low-latency inference. LaWAM runs in 187 ms per action-chunk prediction and achieves up to 24x lower wall-clock latency than pixel-space WAMs.

13.
arXiv (CS.CL) 2026-06-16

Building Customer Support AI Agents at 100M-User Scale: An Evaluation-Driven Framework

The rapid rise in LLM capabilities has made AI agents increasingly viable across a broad range of tasks. Among the most promising applications is building production-ready customer-facing agents, a challenge that demands coordinated excellence in evaluation methodology, context engineering, training, and online measurement. Yet these critical pillars are typically developed in isolation, creating blind spots that only surface after deployment. In this paper, we present a unified framework that bridges offline development with online impact for customer support AI agents at Nubank, a company with 100M+ users. Our approach integrates several key components: (1) structured context engineering tailored to customer support agents, (2) systematic human-in-the-loop prompt iteration, (3) rigorous LLM judge evaluation with measured inter-rater agreement and GEPA optimization for consistency, and (4) ideation-to-production validation. A central insight is that evaluation-pipeline quality directly determines iteration velocity. We present results from five production deployments spanning distinct domains: card delivery, debt management, credit-limit support, card management, and product explanation. These deployments deliver consistent customer-satisfaction gains while substantially accelerating iteration. In our card-delivery deployment, large-scale A/B testing yields a 37 percentage-point improvement in AI transactional Net Promoter Score and a 29 percentage-point gain in self-service rate over prior agent variants, alongside a strong correlation between offline simulation metrics and online outcomes, demonstrating that eval-driven development reliably predicts production impact. On most use cases, AI satisfaction reaches within a few percentage points of expert human agents.

14.
arXiv (CS.AI) 2026-06-18

Structured Cognitive Loop for Behavioral Intelligence in Large Language Model Agents (Extended Revision: From Behavioral Architecture to Epistemic Accountability)

Authors:

arXiv:2510.05107v5 Announce Type: replace Abstract: The central challenge for AI agents is not only performance but accountability. Agents that act through opaque prompt sequences may produce correct outputs, but they provide little basis for verifying why an action was permitted, where an error occurred, or how responsibility should be assigned. This paper presents the Structured Cognitive Loop as an architecture for accountable behavior in large language model agents. SCL separates cognition, memory, control, and action into distinct modules. The language model proposes. External memory preserves verified state. A lightweight controller checks preconditions, prevents redundant actions, and authorizes execution before tools are used. We evaluate SCL against ReAct and common LangChain agent variants across travel planning, conditional email drafting, and constraint guided image generation. Across 360 episodes, SCL achieves 86.3 percent task success compared with 70.5 to 76.8 percent for prompt based baselines. It also improves goal fidelity, reduces redundant tool calls, increases reuse of intermediate state, and lowers unsupported assertions. This extended revision situates SCL within a broader architecture of epistemic accountability. Subsequent extensions integrate context aware Human in the Loop control, Pool Gated Retrieval, and the Horizon Warrant Commitment framework. Together these components define an agent architecture in which the model proposes, structure decides, evidence is warranted before use, and human judgment is embedded in the trace rather than imposed after the fact. The result is a foundation for AI agents whose decisions are not only effective but also authorized, inspectable, and accountable.

15.
arXiv (CS.LG) 2026-06-17

Maximin Relative Improvement: Fair Learning as a Bargaining Problem

arXiv:2602.04155v2 Announce Type: replace-cross Abstract: When deploying a single predictor across multiple subpopulations, we propose a fundamentally different approach: interpreting group fairness as a bargaining problem among subpopulations. This game-theoretic perspective reveals that existing robust optimization methods such as minimizing worst-group loss or regret correspond to classical bargaining solutions and embody different fairness principles. We propose relative improvement, the ratio of actual risk reduction to potential reduction from a baseline predictor, which recovers the Kalai-Smorodinsky solution. Unlike absolute-scale methods that may not be comparable when groups have different potential predictability, relative improvement provides axiomatic justification including scale invariance and individual monotonicity. We establish finite-sample convergence guarantees under mild conditions.

16.
arXiv (CS.LG) 2026-06-16

Generative Molecular Design with Steerable and Granular Synthesizability Control

arXiv:2505.08774v2 Announce Type: replace-cross Abstract: Designing molecules that are both property-optimal and readily synthesizable is a central challenge in drug discovery. Existing works that do consider synthesizability can jointly output predicted synthesis routes for generated molecules. However, there has been minimal attention in addressing the ease of synthesis and with flexibility to incorporate desired reaction constraints. On the other hand, virtual screening searches for commercially available compounds, but imposes challenges when scaling to ultra-large (billion-size and beyond) chemical spaces. Here, we propose a generative design framework that unifies synthesis-constrained molecular design and ultra-large-scale virtual screening through steerable and granular synthesizability control. Generated molecules satisfy arbitrary multi-parameter optimization objectives with predicted synthesis routes satisfying mix-and-match constraints: including or avoiding certain reactions, incorporating specific building blocks, and minimizing synthesis route length. In an end-to-end in-house campaign targeting BRD4, we designed molecules synthesizable with specific selected reactions and building blocks, synthesized all six selected compounds, and identified two micromolar binders. We further demonstrate that reaction control enables efficient navigation of ultra-large make-on-demand chemical spaces to identify property-optimal candidates. By applying our framework to Chemspace's Freedom 4.0 make-on-demand space (142 billion molecules), we generated ~320k molecules (0.00023% of the library) on a single consumer-grade GPU (with only 8 GB GPU memory) and identified a micromolar Wee1 binder amongst 60 synthesized candidates. The single unified framework thus enables generating novel synthesizable molecules and retrieving catalogue-ready candidates, offering a flexible solution to mitigating the synthesizability bottleneck.

17.
arXiv (CS.LG) 2026-06-19

Multi-Task Bayesian In-Context Learning

arXiv:2606.20538v1 Announce Type: new Abstract: Bayesian predictive inference provides a principled framework for uncertainty quantification, data efficiency, and robust generalization. However, exact inference is often intractable, and scalable approximations may remain computationally expensive or require restrictive modeling assumptions that degrade predictive performance. Prior-Data Fitted and in-context models have recently emerged as an amortized alternative by learning to map datasets directly to predictive distributions, but existing approaches are tightly coupled to the support of the training prior and lack explicit mechanisms for adapting to new priors at test time, resulting in limited robustness under distribution shift. We introduce a multi-task in-context learning framework for amortized hierarchical Bayesian predictive inference that explicitly represents prior information as a prefix of in-context datasets. A transformer trained on sequences of prior and target tasks learns to adapt its predictions across families of priors. On a suite of evaluations with increasing difficulty, including out-of-meta-distribution priors and priors with high-dimensional latent structures, our method matches oracle Bayesian predictors while being orders of magnitude faster. We further demonstrate its practical relevance on a real-world spatiotemporal temperature prediction benchmark. Code is available at https://github.com/martianmartina/multi-task-bayesian-icl/.

18.
medRxiv (Medicine) 2026-06-12

Does the method matter? Evaluating the effectiveness, efficiency and ease of hearing-aid gain self-adjustment

In conventional hearing-aid personalisation, clinicians cannot hear what their patients hear, and patients cannot often reliably detect or describe what they hear. Self-adjustment avoids this issue but requires user controls that adjust hearing-aid signal processing parameters to be effective, efficient and easy. In this study, we explored (a) the roles of interface complexity and stimulus type in the self-adjustment of hearing-aid gain, and (b) how well individuals can adjust one sound to match another to assess the same interfaces and stimuli. Adult hearing-aid users with mild to moderate symmetrical sensorineural hearing loss repeatedly adjusted the gain (a) to their preference from individual prescription (n = 41) and (b) to match their previous preferences from a random starting point (n = 32) using three interfaces representing different bass/mid/treble configurations and three stimuli (music, speech and speech-in-noise). The large interindividual variability in self-adjusted gains clustered into three patterns of deviation from initial prescription: increased relative bass, overall gain reduction, and close to initial prescription. There were no substantial effects of interface nor stimulus on self-adjustment reliability (median {sigma} = 2.8 dB), whereas absolute sound-matching error increased with increasing interface complexity and centre frequency. Neither individual matching accuracy nor questionnaire responses predicted either self-adjusted gains or reliability. Overall, these results show that many - but not all - hearing-aid users can adjust gains with reasonable reliability, and while it can be difficult to predict the behaviour from the individual, the individual applies a similar self-adjustment behaviour across different interfaces and stimuli.

19.
medRxiv (Medicine) 2026-06-22

Between Patterns and Predictions: Interpretable Latent EEG Representations for Clinical Insights

Electroencephalography (EEG) captures rich brain dynamics, yet in clinical practice this complexity is often reduced to simplified summaries or categorical labels, limiting its interpretability for decision-making. We tested the hypothesis that a pretrained latent embedding framework, the Universal Map of EEG (UM-EEG), can preserve clinically meaningful structure across heterogeneous datasets and provide a generalizable representation of brain states. We applied UM-EEG, without retraining, to three independent cohorts spanning distinct clinical contexts: long-term EEG recordings from cardiac arrest patients (n = 576), subarachnoid hemorrhage (n = 100), and routine clinical EEG recordings containing physiological and pathological patterns (n = 141). EEG segments were projected into a shared 128-dimensional space anchored by expert-derived reference states, including wakefulness, sleep stages, ictal-interictal continuum activity, and burst suppression. Across datasets, favorable outcome or physiological recordings were consistently located closer to healthy reference states, whereas poor outcome and pathological recordings shifted toward pathological regions of the embedding space. Trajectory-derived geometric and temporal features discriminated outcome in cardiac arrest (ROC-AUC 0.83) and subarachnoid hemorrhage (ROC-AUC 0.76), and distinguished physiological from pathological routine EEGs (ROC-AUC 0.93). In routine EEG, similarity relationships derived from embedding trajectories correlated with those derived from structured clinical reports, indicating that the latent space recapitulates clinically relevant organization. These findings show that a fixed, semantically structured EEG embedding generalizes across etiologies and recording settings, enabling prognostic stratification and contextual interpretation while preserving the relational structure of brain states.

21.
medRxiv (Medicine) 2026-06-17

The Unreliable Judges: Assessing Reproducibility and Self-Preference Bias of LLMs as Free-Text Evaluators

Large Language Models (LLMs) are transforming clinical practice and research, but their adoption requires rigorous evaluation. While human assessment is ideal, its cost has driven the widespread use of LLMs as evaluators. We introduce an open-source reciprocal framework comparing 71 human experts against six LLMs. AI evaluators show a strong self-preference bias, yet neither group reliably identified whether a response was human- or AI-generated. AI scores correlated with surface features such as length and lexical diversity, whereas human scores did not. By probing the evaluator's hidden states and applying targeted steering, we show that verbosity is a major causal driver of the bias. Moreover, shuffling question-response pairings shows that long responses keep high scores even when they no longer answer the question, whereas short ones do not, demonstrating that AI judges reward verbosity largely independently of content alignment. Finally, API-based and batch inference inflate stochasticity, underscoring the need for controlled deployment.

22.
arXiv (CS.LG) 2026-06-17

Learning Credal Ensembles via Distributionally Robust Optimization

arXiv:2602.08470v3 Announce Type: replace Abstract: Credal predictors are models that are aware of epistemic uncertainty and produce a convex set of probabilistic predictions. They offer a principled way to quantify predictive epistemic uncertainty (EU) and have been shown to improve model robustness in various settings. However, most state-of-the-art methods mainly define EU as disagreement caused by random training initializations, which mostly reflects sensitivity to optimization randomness rather than uncertainty from deeper sources. To address this, we define EU as disagreement among models trained with varying relaxations of the i.i.d. assumption between training and test data. Based on this idea, we propose CreDRO, which learns an ensemble of plausible models through distributionally robust optimization. As a result, CreDRO captures EU not only from training randomness but also from meaningful disagreement due to potential distribution shifts between training and test data. Empirical results show that CreDRO consistently outperforms existing credal methods on tasks such as out-of-distribution detection across multiple benchmarks and selective classification in medical applications.

23.
arXiv (CS.CV) 2026-06-11

FronTalk: Benchmarking Front-End Development as Conversational Code Generation with Multi-Modal Feedback

We present FronTalk, a benchmark for front-end code generation that pioneers the study of a unique interaction dynamic: conversational code generation with multi-modal feedback. In front-end development, visual artifacts such as sketches, mockups and annotated creenshots are essential for conveying design intent, yet their role in multi-turn code generation remains largely unexplored. To address this gap, we focus on the front-end development task and curate FronTalk, a collection of 100 multi-turn dialogues derived from real-world websites across diverse domains such as news, finance, and art. Each turn features both a textual instruction and an equivalent visual instruction, each representing the same user intent. To comprehensively evaluate model performance, we propose a novel agent-based evaluation framework leveraging a web agent to simulate users and explore the website, and thus measuring both functional correctness and user experience. Evaluation of 20 models reveals two key challenges that are under-explored systematically in the literature: (1) a significant forgetting issue where models overwrite previously implemented features, resulting in task failures, and (2) a persistent challenge in interpreting visual feedback, especially for open-source vision-language models (VLMs). We propose a strong baseline to tackle the forgetting issue with AceCoder, a method that critiques the implementation of every past instruction using an autonomous web agent. This approach significantly reduces forgetting to nearly zero and improves the performance by up to 9.3% (56.0% to 65.3%). Overall, we aim to provide a solid foundation for future research in front-end development and the general interaction dynamics of multi-turn, multi-modal code generation. Code and data are released at https://github.com/shirley-wu/frontalk

24.
arXiv (CS.LG) 2026-06-16

An Integrable Token Mixing Layer from the Generalized Yang Baxter Equation

arXiv:2606.15085v1 Announce Type: new Abstract: The YB Mixer is a sequence token mixing layer derived from free fermion and generalized Yang Baxter structures. It applies a core principle from integrable systems where a local algebraic constraint guarantees global computational stability. By using the Ising exchange algebra the mixer creates a free fermionic structure that acts as an exactly norm preserving orthogonal map. This algebra also produces commuting transfer matrices which allow inference to be order free and adaptable to any variable budget. To ensure the model can generalize to longer sequence lengths it uses a spectral circulant generator. This generator maintains the crucial orthogonal and commuting properties of the system. The result is a highly stable and mathematically grounded architecture for sequence processing.

25.
arXiv (CS.CL) 2026-06-11

Pre-AF 13: An Interpretable Atrial Fibrillation Risk Score Mined from Discharge Reports

Background. Atrial fibrillation (AF) is the most prevalent cardiac arrhythmia and a major determinant of prognosis. Established AF risk scores rely on factors (older age, hypertension) nearly ubiquitous among patients with cardiovascular disease (CVD), offering limited stratification in this high-risk group. Most target long-term (5-10 year) rather than medium-term prediction. We developed interpretable ML models predicting AF risk over a 24-month and entire follow-up horizon in CVD patients using routinely collected hospital data. Methods. Single-center retrospective study of electronic health records from the National Research Cardiology Center (Russia) for patients aged >=18 with CVD but without pre-existing AF, hospitalized more than once between January 2012 and May 2019. A custom NLP pipeline transformed unstructured discharge reports into 73 structured features, combining a rule-based parser with transformer-based NER. Using LightAutoML we built a full model (73 features), a simple model (reduced subset), and a linear model for a bedside risk score. Performance was assessed by ROC AUC, compared with CHARGE-AF, C2HEST, MHS, and HAVOC, and interpreted via SHAP. Results. Of 80,576 records from 45,000 patients, 17,562 met inclusion criteria; 1,438 (8.19%) developed AF. The full model reached ROC AUC 0.735 (24-month) and 0.696 (entire follow-up); the simple model was nearly identical (0.725, 0.696). All non-linear models outperformed the four clinical risk scores (ROC AUC 0.53-0.64). The simple model uses 13 features and is named Pre-AF 13. SHAP identified age and left atrial volume as dominant predictors. A linear risk score (Pre-AF 9) stratified observed 24-month AF incidence from ~7% to 36%. Conclusion. Interpretable ML models built from routinely collected EHR data identify high-AF-risk CVD patients, outperforming established clinical risk scores.